Our hypothesis is that advanced magnetic resonance imaging (MRI) techniques in combination with the quantitative evaluation of cerebrospinal fluid (CSF) metabolites can be exploited to develop a comprehensive set of biomarkers for assessing disease severity and therapeutic response in children with late infantile neuronal ceroid lipofuscinosis (LINCL). The ultimate goal of our work is to accurately and objectively assess therapeutic efficacy and thereby improve the medical management of LINCL subjects, especially as new treatment options for LINCL become available. Our specific goal is to develop an imaging biomarker panel for assessing LINCL disease severity and therapeutic response at the level of individual substructures in the brain, and compare this with the results of CSF metabolomics. Our Specific Aims are as follows: Aim 1) To develop a brain-region-specific magnetic resonance imaging biomarker panel, including diffusion coefficients, diffusion fractional anisotropy, T2 relaxation times, T2* relaxation times, the volume percentage of CSF, and proton spectroscopic metabolite ratios. We will examine over 30 brain regions in 16 LINCL subjects in a control group at the time of enrollment, and again 18 months later. Aim 2: To develop a CSF metabolite biomarker panel that includes the evaluation of over 1000 compounds, and test its correlation with the MRI biomarkers, clinical indicators of disease severity, and subject age. CSF will be collected from the 16 LINCL subjects in the control group via lumbar puncture performed under anesthesia at one day prior to administration of gene therapy and again 18 months later. Aim 3: To apply the MRI biomarker panel in the evaluation of therapeutic efficacy of AAVrh.10 mediated gene therapy for LINCL. Sixteen subjects in a treatment group will be examined at 5 MRI time points, including at the time of screening, one day before administration of therapy, and again at 6, 12 and 18 months post therapy.